In our R21-funded pilot work, we developed and tested a novel Cognitive Behavioral Therapy (CBT) treatment program for individuals with co-occurring alcohol use disorder and anxiety disorder ("co-morbidity"). In stage 1 of the R21, we created the hybrid CBT program by modifying the standard elements and delivery systems of CBT treatment for anxiety disorder to fit the specific problem of co-morbidity among patents undergoing treatment for alcoholism. In stage 2 of the R21, we conducted a clinical trial showing that the hybrid CBT treatment substantially lowered anxiety disorder symptoms and risk for relapse to drinking by 4- month follow-up compared to those patients not receiving the hybrid CBT treatment. We now propose to conduct a controlled clinical trial of the hybrid CBT treatment package (stage 3) in which we will study 60 alcoholism treatment patients (20 each with social phobia, panic disorder and generalized anxiety disorder) in each of two treatment conditions (CBT vs. active placebo control) (N=120). In the CBT condition, three treatment units will each be the focus of two therapy sessions and two "study hall" sessions (12 sessions total). In study halls, subjects are proctored as they review study materials and practice coping skills taught in the therapy sessions. Control subjects undergo relaxation training (RT) on "therapy" days and will read about and practice RT on "study hall" days. (RT is structured and credible but has little effect on anxiety disorder or drinking outcomes.)The study interventions and alcoholism treatment-as-usual will run in parallel; following which, subjects will undergo assessments at post-treatment, four-months, and 12-months. The central prediction is that the$ CBT treatment group will demonstrate less relapse at follow-up than placebo controls and we will also explore whether this effect is mediated by anxiety symptom reduction and whether the CBT treatment effects differ for (i.e., are moderated by) the anxiety disorder subtype. Confirmation of the study's central hypothesis would support the conclusion that the hybrid CBT treatment package could significantly mitigate risk of relapse to drinking among the many alcohol-disordered patients with co-morbid anxiety disorder.